The SiFive Intelligence XM Series is designed for accelerating high performance AI workloads. This is the first IP from SiFive to include a highly scalable AI matrix engine, which accelerates time to market for semiconductor companies building powerful system on chip solutions. SiFive understands that companies need a flexible and scalable hardware solution to maximize AI software investment.

Featuring four X-Cores per cluster, a cluster can deliver 16 TOPS (INT8) or 8 TFLOPS (BF16) per GHz. There is 1TB/s of sustained memory bandwidth per Intelligence XM Series cluster, with the clusters being able to access memory via a high bandwidth port or via a CHI port for coherent memory access. This flexible design supports multiple host and memory configurations. SiFive envisions the creation of systems incorporating no host CPU or ones based on RISC-V, x86 or Arm.

The SiFive Intelligence XM Series is ideal for edge IoT, consumer devices, next generation electric and/or autonomous vehicles, datacenters, and beyond. The Intelligence XM series has massive scalability for customers to build up to industry leading levels of performance, whether they are using AI workloads in the datacenter or are doing data flow management, object detection, speech and recommendation processing, or other types of AI applications. By integrating scalar, vector, and matrix engines, SiFive Intelligence XM Series customers can take advantage of very efficient memory bandwidth. SiFive Intelligence XM Series also continues the company's legacy of offering extremely high performance per watt for compute-intensive applications.